Role of Transthoracic Ultrasound in Differentiation of the Causes of Pleural Thickening Khaled M

260 Original article

Role of transthoracic ultrasound in differentiation of the causes of pleural thickening Khaled M. Kamel, Yasmine H. El-Hinnawy

Objectives Pleural thickening is defined as the increase in Results There was a statistically significant relation between thickness of the pleura of more than 3 mm and can be caused the distribution either localized, diffuse, unilateral, or bilateral; by a wide range of diseases, either nonmalignant or the surface of the thickness; invasion of chest wall or malignant. Thoracic ultrasound has high sensitivity in diaphragm; the echogenicity; vascularity of the pleural assessing the pleura. thickness; and the presence of pleural effusion and its pattern on one hand and the diagnosis of pleural thickening on the Aim The aim of this study was to assess the role of thoracic other hand. There was insignificant statistical difference ultrasound in differentiation of the causes of pleural between pleural mesothelioma and pleural metastatic cases, thickening. and also there was insignificant statistical difference between Design A prospective study included 48 patients selected tuberculous and nonspecific infection cases. from the inpatient Chest Department, Kasr Al-Ainy Hospital, Conclusion The transthoracic ultrasound had a very good from January 2016 till October 2017. Patients diagnosed as predilection for the diagnosis of pleural thickening etiology having pleural thickening underwent thoracic ultrasound as whether malignant or nonmalignant. well as ultrasound-guided pleural biopsy by Tru-cut needle. Egypt J Bronchol 2018 12:260–265 Descriptive data were obtained including age and sex of the © 2018 Egyptian Journal of Bronchology patients. Thoracic ultrasound was done for the side of pleural thickening. The distribution of pleural thickness, either Egyptian Journal of Bronchology 2018 12:260–265 localized or diffuse; the surface; invasion of chest wall or Keywords: benign, malignant, pleural thickening, ultrasound diaphragm; the echogenicity and vascularity; and the presence of pleural effusion and its pattern were determined. Chest Department, Faculty of Medicine, Cairo University, Giza, Egypt The patients were classified into two main groups: Correspondence to Yasmine H. El-Hinnawy, MD, 16 El-Tayaran Street, Nasr nonmalignant (subclassified as tuberculous and nonspecific City, Cairo, 11759, Egypt. Tel: +20 122 233 7683; e-mail: [email protected] infection) and malignant cases (subclassified as mesothelioma and metastatic cases). Received 2 November 2017 Accepted 4 December 2017

Introduction Inclusion criteria Pleura is a serous membrane. It consists of parietal pleura Patients diagnosed as having pleural thickening and visceral pleura. Pleural thickening is defined as the and undergoing transthoracic ultrasound as well as increase inthickness ofthe pleuraofmorethan3 mm[1]. ultrasound-guided pleural biopsy by Tru-cut needle were included. There is quite a wide range of diseases that can cause pleural thickening. It can be either benign or malignant Exclusion criteria [2]. Benign pleural thickening can occur in pleuro- The exclusion criteria were the presence of pulmonary infection by either a specific organism as contraindication for pleural biopsy. in tuberculosis [3] or a nonspecific organism causing empyema [4]. Transthoracic ultrasound is now Descriptive data of the study population were obtained considered a gold standard radiological technique in including age and sex of the patients. studying the pleural diseases such as pleural thickening [5] with or without pleural effusion of different etiologies Examination was performed using a real-time [6]. It is the most sensitive and safe technique in detecting ultrasound scanner (Hitachi EUB-7000 with minimal pleural thickening or effusion [7]. 3.5 MHz convex probe transducer and 13 MHz linear probe transducer, Hitachi, Tokyo, Japan). All The aim of this study was to assess the role of transthoracic patients were examined in an upright sitting position or ultrasound in differentiation of the causes of pleural the lateral decubitus position. thickening.

Patients and methods This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 A prospective study included 48 patients who were License, which allows others to remix, tweak, and build upon the work selected from the Chest Department inpatients, Kasr non-commercially, as long as appropriate credit is given and the new Al-Ainy Hospital, from January 2016 till October 2017. creations are licensed under the identical terms.

We searched by transthoracic ultrasound for side Table 1 Descriptive data of the study population in the two of pleural thickening. The distribution of pleural main groups thickness either localized or diffuse, the surface, Nonmalignant Malignant P-value group (N=22) group (N=26) invasion of either chest wall or diaphragm, the echogenicity, vascularity, and the presence of pleural Age (years) effusion and its pattern were evaluated. Mean 44.045 57.423 0.0058* SD 18.574 11.503 Sex We classified the patients into two main groups Male 15 17 0.8376 (nonmalignant and malignant). The nonmalignant Female 7 9 cases were subclassified as tuberculous and nonspecific Pleural thickening (mm) infection, whereas malignant cases were subclassified as Mean 8.995 21.826 0.0001* mesothelioma and metastatic cases. SD 3.220 14.022 Side Unilateral 18 26 0.0231* Statistical methods Bilateral 4 0 Coding and entering of the data was done using the Distribution statistical package statistical package for the social Localized 2 9 0.0360* sciences, version 23 (SPSS; SPSS Inc., Chicago, Diffuse 20 17 Illinois, USA). Surface Smooth 22 0 0.00001* Data were summarized using mean and SD for Irregular 0 26 Invasion quantitative variables and frequencies (number of cases). Yes 0 19 0.00001* No 22 7 Comparisons between groups were done using unpaired Echogenicity t-test [8]. Echogenic 22 4 0.00001* Hypoechogenic 0 22 Comparison of numerical variables between more Vascularity than two groups was done using one-way analysis Avascular 3 0 0.0021* of variance test with post-hoc multiple two-group Scanty 16 8 Vascular 3 18 comparisons in normal data [9]. Statistical P Pleural effusion significant was considered when value was less No 0 10 0.0073* than 0.05. Mild 10 5 Moderate 12 7 Massive 0 4 Results Pleural effusion pattern The descriptive data among the two main groups No 0 10 0.0078* (nonmalignant and malignant) are shown in Table 1. Complex 20 12 A total of 48 patients formed the study population. septated Complex 24 nonseptated The nonmalignant cases were 22, whereas the *Statistically significant. malignant cases were 26. The mean±SD age was 44.045±18.574 in nonmalignant group, and it was with statistically significant difference (P=0.00001). 57.423±11.503 in malignant group, with statistically There was no invasion to the chest wall or significant difference (P=0.0058). The sex distribution diaphragm in all nonmalignant cases whereas seven did not show any statistical significance. The pleural malignant cases showed invasion (Fig. 3), and it was thickening mean±SD was 8.995±3.22 in nonmalignant statistically significant (P=0.00001). cases, and it was 21.826±14.022 in malignant cases, with statistically significant difference (P=0.0001). All All the nonmalignant cases showed echogenic pleural cases were unilateral in the malignant group, and a thickening, and 22 of the malignant cases showed majority in the nonmalignant group, with statistically hypoechogenic pleural thickening, with statistically significant difference (P=0.0231). significant difference (P=0.00001).

The distribution was statistically significant (P=0.036). The pleural thickening in the nonmalignant cases can The pleural surface was smooth in all nonmalignant be avascular (3/22), scanty (16/22), and vascular (3/22) cases (Figs 1 and 2) and irregular in all malignant cases, unlike malignant cases (18/26), which it was vascular 262 Egyptian Journal of Bronchology, Vol. 12 No. 2, April-June 2018

Figure 1

A case of chronic nonspecific pleural inflammation and thickening. (a) Computed tomography chest mediastinal window showed right pleural effusion with pleural thickening. (b) High-frequency linear probe transthoracic ultrasound B-mode image showed complex septated pleural effusion with diffuse smooth pleural thickening. (c) High-frequency linear probe transthoracic ultrasound B-mode image with Doppler study showed poor vasculature in pleural thickening.

Figure 2

A case of tuberculous pleural inflammation and thickening. (a) Computed tomography chest mediastinal window showed bilateral pleural effusion with pleural thickening. (b) High-frequency linear probe transthoracic ultrasound B-mode image showed complex septated pleural effusion with diffuse smooth pleural thickening. (c) High-frequency linear probe transthoracic ultrasound B-mode image with Doppler study showed good vasculature in pleural thickening.

Figure 3

A case of right-sided mesothelioma. (a) Computed tomography chest mediastinal window showed right irregular pleural thickening. (b) High- frequency linear probe transthoracic ultrasound B-mode image showed irregular pleural thickening with sites of chest wall invasion. (c) High- frequency linear probe transthoracic ultrasound B-mode image with Doppler study showed poor vasculature in pleural thickening.

(Fig. 4) with no avascular cases, with statistically comparison with malignant cases, which show significant difference (P=0.0021). absence of pleural effusion in 10 cases.

All nonmalignant cases were associated with pleural There was statistically significant difference between effusion either mild or moderate amount, with nonmalignant and malignant cases regarding the statistically significant difference (P=0.0073) in pattern of pleural effusion (P=0.0078). Transthoracic ultrasound in pleural thickening Kamel and El-Hinnawy 263

Figure 4

A case of left pleural effusion with pleural metastasis. (a) Computed tomography chest mediastinal window showed left massive pleural effusion with pleural deposits. (b) High-frequency linear probe transthoracic ultrasound B-mode image showed irregular pleural thickening and nodulations. (c) High-frequency linear probe transthoracic ultrasound B-mode image with Doppler study showed good vascularized pleural nodule.

The descriptive analysis within the nonmalignant cases The pleural thickening was unilateral in all cases of the is shown in Table 2. The nonmalignant group was malignant group and majority in nonmalignant group, subclassified into two groups: one with tuberculous with statistically significant difference (P=0.0231), so infection group and another with nonspecific infection bilateral pleural thickening is more with nonmalignant group. There was statistically significant difference etiology. This goes with the fact that bilateral regarding age and sex distribution between the two malignant pleural affection is uncommon [12]. subgroups (P=0.0001 and 0.0467, respectively). Regarding the distribution of pleural thickening, localized There was no statistical significance regarding the pleural thickening was more common in malignant group thickness of the pleura, side, distribution, surface, with statistically significant difference (P=0.036). invasion, echogenicity, vascularity, and pleural effusion between tuberculous infection group and The pleural surface was smooth in all nonmalignant the nonspecific infection group. cases and irregular in all malignant cases, with statistically significant difference (P=0.00001). This The descriptive analysis within the malignant cases is is agreed by Qureshi et al. [13] who stated that the showninTable3.Themalignantgroup was subclassified surface in malignant cases is irregular. Moreover, Kao into two groups: one with mesothelioma and another et al. [14] considered pleural nodularity as a sign of group with metastasis to the pleura. There was no malignancy, which agreed with our study. statistical significance regarding the age, sex, thickness of the pleura, side, distribution, surface, invasion, There was no invasion to the chest wall or diaphragm in echogenicity, vascularity, and pleural effusion between all nonmalignant cases whereas seven malignant cases mesothelioma group and the metastatic group. showed invasion of either chest wall or diaphragm, and it was statistically significant (P=0.00001). This finding agreed with Wernacke [15]. Discussion Thoracic ultrasound is a very important radiological All the nonmalignant cases showed echogenic pleural technique in assessing the nature of pleural opacities thickening and 22 of the malignant cases showed and effusions [10] and it also helps in distinguishing hypoechogenic pleural thickening, with statistically pleural thickening from minimal pleural effusion significant difference (P=0.00001). This difference [5,11]. may be owing to presence of more fibrous tissue in the thickened pleura of nonmalignant cases. This also The mean±SD of the pleural thickness was 8.995± agrees with Dietrich et al. [16], who described that 3.22 in nonmalignant cases and 21.826±14.022 in postinflammatory pleural thickening is echogenic. malignant cases, with statistically significant difference (P=0.0001). This can be explained by the more The pleural thickening in the nonmalignant cases can be proliferation of malignant tissues. Tsai and Yang [5], avascular (3/22), scanty (16/22), and vascular (3/22) unlike considered that there was pleural thickening if it malignant cases, where it was vascularize in 18/26, with no measured more than 3 mm. avascular case, with statistically significant difference 264 Egyptian Journal of Bronchology, Vol. 12 No. 2, April-June 2018

Table 2 Descriptive data of the study population in the Table 3 Descriptive data of the study population in the nonmalignant group malignant group Tuberculous Nonspecific P-value Mesothelioma Metastatic P-value infection group infection group group (N=22) group (N=4) (N=9) (N=13) Age (years) Age (years) Mean 56.545 62.25 0.6523 Mean 28 55.153 0.00001* SD 9.179 21.823 SD 10.331 14.345 Sex Sex Male 14 3 0.6603 Male 4 11 0.0467* Female 8 1 Female 5 2 Pleural thickening (mm) Pleural thickening (mm) Mean 23.113 14.75 0.32 Mean 8.2 9.546 0.4087 SD 14.165 12.446 SD 4.284 2.260 Site Side Unilateral 22 4 0.9999 Unilateral 6 12 0.1252 Bilateral 0 0 Bilateral 3 1 Distribution Distribution Localized 5 4 0.0028* Localized 1 1 0.7838 Diffuse 17 0 Diffuse 8 12 Surface Surface Smooth 0 0 1 Smooth 9 13 0.9999 Irregular 22 4 Irregular 0 0 Invasion Invasion Yes 17 2 0.2579 Yes 0 0 1 No 5 2 No 9 13 Echogenicity Echogenicity Echogenic 3 1 0.5622 Echogenic 9 13 0.9999 Hypoechogenic 19 3 Hypoechogenic 0 0 Vascularity Vascularity Avascular 0 0 0.14720 Avascular 1 2 0.9021 Scanty 8 0 Scanty 6 10 Vascular 14 4 Vascular 2 1 Pleural effusion Pleural effusion No 10 0 0.32625 No 0 0 0.60670 Mild 4 1 Mild 3 7 Moderate 5 2 Moderate 6 6 Massive 3 1 Massive 0 0 Pleural effusion pattern Pleural effusion pattern No 10 0 0.06345 No 0 0 0.63152 Complex 10 2 Complex 812 septated septated Complex 22 Complex 11 nonseptated nonseptated *Stastically significant. *Statistically significant. to nonspecific infection (P=0.00001). Moreover, (P=0.0021). The high vasculature in thickened pleura of pleural thickening owing to nonspecific infection was malignant cases is mostly owing to neoangiogenesis which more common in males, with statistically significant is a characteristic of malignancy. Koh et al.[17]described difference (P=0.0467). Otherwise no other feature can the vascularity of malignant pleural masses as being differentiate between pleural thickening due to tortious and irregular. Not all vascularized pleural tuberculosis or nonspecific infection. thickening is considered to be malignant. Malignant issues depend on blood vessels for their growth, so it is The descriptive analysis within the nonmalignant either of the host or by neovascularization [18]. cases showed that the localized pleural thickening was the usual presentation of pleural metastasis The descriptive analysis within the nonmalignant cases (P=0.0028) in comparison with mesothelioma showed that tuberculous pleural thickening was more which may present with diffuse or localized pleural common in relatively younger age patients in affection. Otherwise we cannot differentiate between comparison with cases with pleural thickening owing pleural thickening owing to mesothelioma and pleural Transthoracic ultrasound in pleural thickening Kamel and El-Hinnawy 265

thickening because of pleural metastasis except by 6 Soni NJ, Franco R, Velez MI, Schnobrich D, Dancel R, Restrepo MI, et al. Ultrasound in the diagnosis & management of pleural effusions. J Hosp tissue pathology. Med 2015; 10:811–816. 7 Mercer RM, Psallidas I, Rahman NM. Ultrasound in the management of pleural disease. Expert Rev Respir Med 2017; 11:323–331. Conclusion 8 Chan YH. Biostatistics 102: quantitative data−parametric & non- – The transthoracic ultrasound has a very good parametrictests. Singapore Med J 2003; 44:391 396. 9 Rosner B. Fundamentals of biostatistics. Belmont: Duxbury Press 2006. predilection for the diagnosis of pleural thickening pp. 557–558. etiology whether malignant or nonmalignant. 10 Rosenberg ER. Ultrasound in the assessment of pleural densities. Chest 1983; 84:283–285. 11 Kalokairinou-Motogna M, Maratou K, Paianid I, Soldatos T, Antipa E, Financial support and sponsorship Tsikkini A, et al. Application of color Doppler ultrasound in the study of – Nil. small pleural effusion. Med Ultrason 2010; 12:12 16. 12 Bugalho A, Ferreira D, Dias SS, Schuhmann M, Branco JC, Marques Gomes MJ, et al. The diagnostic value of transthoracic Conflicts of interest ultrasonographic features in predicting malignancy in undiagnosed pleural effusions: a prospective observational study. Respiration There are no conflicts of interest. 2014; 87:270–278. 13 Qureshi NR, Rahman NM, Gleeson FV. Thoracic ultrasound in the diagnosis of malignant pleural effusion. Thorax 2009; 64:139–143. References 14 Kao S, Yen A, Nakanote K, Brouha S. Silver lining: imaging manifestations 1 Sureka B, Thukral BB, Mittal MK, Mittal A, Sinha M. Radiological review of of pleural pathology. Appl Radiol 2016; 45:9–23. pleural tumors. Indian J Radiol Imaging 2013; 23:313–320. 15 Wernecke K. Ultrasound study of the pleura. Eur Radiol 2006; 2 Downer NJ, Ali NJ, Au-Yong LTH. Investigating pleural thickening. BMJ 10:1515–1523. 2013; 346:e8376. 16 Dietrich CF, Mathis G, Cui XW, Ignee A, Hocke M, Hirche TO. 3 Vorster MJ, Allwood BW, Diacon AH, Koegelenberg CFN. Tuberculous Ultrasound of the pleurae and lungs. Ultrasound Med Biol 2015; pleural effusions: advances and controversies. J Thorac Dis 2015; 41:351–365. 7:981–991. 17 Koh DM, Burke S, Davies N, Padley SPG. Transthoracic US of the chest: 4 Karkhanis VS, Joshi JM. Pleural effusion: diagnosis, treatment, and clinical uses and applications. Radiographics 2002; 22:e1. management. Emerg Med 2012; 4:31–52. 18 Nagy JA, Chang SH, Dvorak AM, Dvorak HF. Why are tumour blood 5 Tsai TH, Yang PC. Ultrasound in the diagnosis and management of pleural vessels abnormal and why is it important to know?. Br J Cancer 2009; disease. Curr Opin Pulm Med 2003; 9:282–290. 100:865–869.